NVIDIA A100 PCIe 80 GB vs NVIDIA A100 SXM4 80 GB

Comparison of NVIDIA A100 PCIe 80 GB with 80 GB HBM2e and 6,912 cores vs NVIDIA A100 SXM4 80 GB with 80 GB HBM2e and 6,912 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

NVIDIA A100 PCIe 80 GB

NVIDIA A100 PCIe 80 GB

MI250 MI250
Instinct MI300X Instinct MI300X
RX 7900 XTX RX 7900 XTX

NVIDIA A100 SXM4 80 GB

NVIDIA A100 SXM4 80 GB

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Additional

Memory

Memory Size

80 GB 80 GB

Memory Type

HBM2e HBM2e

Memory Bandwidth

1.94 TB/s 2.04 TB/s

Memory Bus Width

5,120 бит 5,120 бит

ML Performance

FP16 (Half Precision)

77.97 TFLOPS 77.97 TFLOPS

BF16 (Brain Float)

311.84 TFLOPS 311.84 TFLOPS

TF32 (TensorFloat)

155.92 155.92

Compute Power

FP32 (Single Precision)

19.49 TFLOPS 19.49 TFLOPS

FP64 (Double Precision)

9.746 TFLOPS 9.746 TFLOPS

CUDA Cores

6,912 6,912

RT Cores

No No

Architecture & Compatibility

GPU Architecture

Ampere Ampere

SM (Streaming Multiprocessor)

108 108

PCIe Version

PCIe 4.0 x16 PCIe 4.0 x16

ML Software Support

CUDA Version

8.0 8.0

Clocks & Performance

Base Clock

1,065 1,275

Boost Clock

1,410 1,410

Memory Clock

1,512 1,593

Power Consumption

TDP/TGP

🔥 -25% 300 W
400 W

Recommended PSU

🔥 -12% 700 W
800 W

Power Connector

8-pin EPS None

Rendering

Texture Units (TMU)

432 432

ROP

No No

L2 Cache

80 MB 40 MB

Additional

Slots

Dual-slot OAM Module

Release Date

June 28, 2021 Nov. 16, 2020

Display Outputs

No outputs
No outputs

Renting is cheaper than buying